15Low
This skill describes how to dispatch parallel agents with
Behavior Analysis
Mismatch detected — The skill's stated purpose implies it performs the action of dispatching parallel agents. However, its actual content is purely instructional, describing *how* to dispatch agents using an internal `Task()` function, rather than being an executable component that performs the dispatching itself.
Claims to do
Dispatching Parallel Agents: You delegate tasks to specialized agents with isolated context. By precisely crafting their instructions and context, you ensure they stay focused and succeed at their task. They should never inherit your session's context or history — you construct exactly what they need. This also preserves your own context for coordination work.
Actually does
This skill is a descriptive guide outlining a strategy for an AI agent to decompose complex problems into independent sub-tasks. It illustrates how to use an internal `Task()` function to dispatch these sub-tasks to parallel sub-agents, providing examples of prompt structures and scenarios. It does not execute any external tools, access external data, run system commands, or contact URLs itself.
Skill Info
Nameobra/superpowers/dispatching-parallel-agents
Registrygithub
Versionc4bbe651cb1b
PURLpkg:github/obra/superpowers@c4bbe651cb1b?skill=dispatching-parallel-agents
Stars180,084
Installs54,700
Source
SHA-25676806091c7f9...
SHA-1fb3cf9870113...
MD5af319c60c986...
TLSH1fd1b536b052...
ssdeep192:24EA3ynC...
Size6,441 bytes
Install
Assessments (2)
AI Agent Traps ↗Sub-agent spawning capabilitylowsub agent spawning
The skill explicitly describes and demonstrates the ability to spawn multiple sub-agents concurrently using `Task()`. This capability, if used with attacker-controlled prompts, could lead to resource abuse or coordinated malicious actions.
100% confidenceline 70
Task("Fix agent-tool-abort.test.ts failures")Sub-agent file system read/write capabilitylowcommand execution
Sub-agents are instructed to read test files and fix issues by modifying code, indicating they possess file system read and write capabilities. This capability could be leveraged for unauthorized data access or code injection if the sub-agent's instructions are malicious.
100% confidenceline 80
Read the test file and understand what each test verifies Fix by: ... Fixing bugs in abort implementation if found
Description Mismatchinfodescription mismatch
The skill's stated purpose implies it performs the action of dispatching parallel agents. However, its actual content is purely instructional, describing *how* to dispatch agents using an internal `Task()` function, rather than being an executable component that performs the dispatching itself.
85% confidence
The content consists of markdown text, diagrams, and code examples (e.g., `Task("...")`) that illustrate a pattern, but no executable code within the skill itself that would perform these actions.Version History (4)
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/obra/superpowers/dispatching-parallel-agents)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/obra/superpowers/dispatching-parallel-agents"><img src="https://mondoo.com/ai-agent-security/api/badge/github/obra/superpowers/dispatching-parallel-agents.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/obra/superpowers/dispatching-parallel-agents.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow